The goals of this project are twofold: (1) To determine whether blasts from patients with relapsed or refractory acute leukemia have detectable levels of p-glycoprotein (Pgp), a protein characteristic of the multidrug resistant (MDR) phenotype and (2) To perform a pilot trial of tamoxifen in conjunction with daunorubicin in patients with advanced leukemia. Preliminary data from our laboratory suggest that tamoxifen, in doses that may be achievable in vivo, is able to reverse the MDR phenotype in vitro. We have studied two leukemia cell lines, the lymphoblastic leukemia line CEM and the myeloid leukemia cell line HL-60 along with, respectively, their vinblastine and vincristine-resistant sublines CEM-VBL and HL- 60/RV+. In both MDR resistant sublines, tamoxifen 10 uM is able to increase daunorubicin uptake as measured by fluorescence, in a manner similar to verapamil; at higher doses, a nonspecific effect of membrane permeability is observed. No such effect is observed in the drug sensitive parent cell lines. As the first part of this clinical project, we plan on analyzing Pgp expression in patient samples using flow cytometric analysis with the monoclonal antibody HYB-241 which recognizes an external epitope of the molecule. Patients will then receive high dose tamoxifen for 7 days with daily measurements of serum levels. Daunorubicin 50 mg/m2 will be given on days 5 and 6. Before the start of the protocol and again on day 5, prior to daunorubicin, a marrow sample will be analyzed for in vitro daunorubicin uptake. A repeat bone marrow sample will be analyzed on day 10 to assess cellularity and Pgp expression. Because MDR may play a role in the high relapse rate in acute leukemia, insights gained from studies in which an agent that can potentially reverse that particular phenotype could have important implications for strategies aimed at the treatment of this disease.